CN116259000A - Crime prevention system, crime prevention method, and computer readable recording medium - Google Patents

Crime prevention system, crime prevention method, and computer readable recording medium Download PDF

Info

Publication number
CN116259000A
CN116259000A CN202211577541.1A CN202211577541A CN116259000A CN 116259000 A CN116259000 A CN 116259000A CN 202211577541 A CN202211577541 A CN 202211577541A CN 116259000 A CN116259000 A CN 116259000A
Authority
CN
China
Prior art keywords
person
camera
crime prevention
image data
camera images
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202211577541.1A
Other languages
Chinese (zh)
Inventor
小堀训成
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Toyota Motor Corp
Original Assignee
Toyota Motor Corp
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Toyota Motor Corp filed Critical Toyota Motor Corp
Publication of CN116259000A publication Critical patent/CN116259000A/en
Pending legal-status Critical Current

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G06V20/53—Recognition of crowd images, e.g. recognition of crowd congestion
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • G08B13/19613—Recognition of a predetermined image pattern or behaviour pattern indicating theft or intrusion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/10—Image acquisition
    • G06V10/16—Image acquisition using multiple overlapping images; Image stitching
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/25—Determination of region of interest [ROI] or a volume of interest [VOI]
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/40—Extraction of image or video features
    • G06V10/62—Extraction of image or video features relating to a temporal dimension, e.g. time-based feature extraction; Pattern tracking
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
    • G06V10/75—Organisation of the matching processes, e.g. simultaneous or sequential comparisons of image or video features; Coarse-fine approaches, e.g. multi-scale approaches; using context analysis; Selection of dictionaries
    • G06V10/751—Comparing pixel values or logical combinations thereof, or feature values having positional relevance, e.g. template matching
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/96—Management of image or video recognition tasks
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
    • G06V40/103—Static body considered as a whole, e.g. static pedestrian or occupant recognition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/20—Movements or behaviour, e.g. gesture recognition
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B13/00—Burglar, theft or intruder alarms
    • G08B13/18—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength
    • G08B13/189—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems
    • G08B13/194—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems
    • G08B13/196—Actuation by interference with heat, light, or radiation of shorter wavelength; Actuation by intruding sources of heat, light, or radiation of shorter wavelength using passive radiation detection systems using image scanning and comparing systems using television cameras
    • G08B13/19602—Image analysis to detect motion of the intruder, e.g. by frame subtraction
    • G08B13/19608—Tracking movement of a target, e.g. by detecting an object predefined as a target, using target direction and or velocity to predict its new position
    • G—PHYSICS
    • G08—SIGNALLING
    • G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
    • G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • H04N7/181—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast for receiving images from a plurality of remote sources
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30196—Human being; Person
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30232—Surveillance

Landscapes

  • Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Multimedia (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Health & Medical Sciences (AREA)
  • General Health & Medical Sciences (AREA)
  • Computing Systems (AREA)
  • Artificial Intelligence (AREA)
  • Databases & Information Systems (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Human Computer Interaction (AREA)
  • Social Psychology (AREA)
  • Psychiatry (AREA)
  • Signal Processing (AREA)
  • Business, Economics & Management (AREA)
  • Emergency Management (AREA)
  • Alarm Systems (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Emergency Alarm Devices (AREA)
  • Image Analysis (AREA)

Abstract

The invention discloses a crime prevention system, a crime prevention method, and a computer-readable recording medium. Suspicious activity of a person captured by a surveillance camera is determined. The crime prevention system monitors the actions of persons included in a plurality of camera images continuously captured in time series by the monitoring camera. The crime prevention system detects person areas of persons included in the plurality of camera images, respectively, and performs a tracking process of recognizing the persons included in the plurality of camera images in accordance with a time series based on the person image data included in the person areas. When the recognition by the tracking process is changed to fail in the middle, the person determined to be a suspicious action is included in the camera image. If it is determined that there is a suspicious activity, the output device may notify the alarm information.

Description

Crime prevention system, crime prevention method, and computer readable recording medium
Technical Field
The present disclosure relates to a crime prevention system that determines suspicious actions of persons contained in camera images, a crime prevention method, and a technique of a computer-readable recording medium having a crime prevention program recorded thereon.
Background
Patent document 1 discloses a technique related to an image distribution system for distributing an image captured by a monitoring camera or the like. The image distribution system of this technology includes a camera and a distribution device. The issuing device is provided with a personal authentication database. The distribution device determines a person (for example, a child) reflected on the camera by referring to a personal authentication database using data of the camera image. The determined person data is transmitted to the receiving apparatus via the network together with the camera image data.
Prior art literature
Patent literature
Patent document 1: japanese patent laid-open No. 2009-225398
Disclosure of Invention
Camera images of persons captured by a surveillance camera or the like are considered for crime prevention. There is a possibility that a person performing a suspicious action is included in a camera image captured by a monitoring camera or the like. One of the suspicious actions is, for example, an action of suddenly changing the appearance of clothing or the like in order to evade tracking. Such suspicious actions can be useful for crime prevention if such suspicious actions can be determined from camera images captured by a monitoring camera.
The present disclosure has been made in view of the above-described problems, and an object thereof is to provide a technique capable of specifying suspicious actions of a person imaged by a surveillance camera.
In order to achieve the above object, the present disclosure provides a crime prevention system that monitors actions of persons included in a plurality of camera images continuously captured by a monitoring camera in accordance with a time series. The crime prevention system is provided with: a memory storing a plurality of camera images; and a processor for processing the plurality of camera images stored in the memory. The processor is configured to: tracking processing of detecting person areas of persons included in the plurality of camera images, respectively, and recognizing the persons included in the plurality of camera images in accordance with a time series based on the person image data included in the person areas; and suspicious activity determination processing for determining that the person of the suspicious activity is included in the camera image when the recognition based on the tracking processing is changed to fail in the middle.
In the crime prevention system according to the present disclosure, the plurality of camera images may include a first camera image and a second camera image subsequent to the first camera image, and template image data including personal image data of a personal area of the first camera image may be stored in the memory, and the tracking processing may be configured to compare the personal image data included in the personal area of the second camera image with the template image data using the neural network model, thereby identifying a person of the second camera image.
In the crime prevention system according to the present disclosure, the processor may be configured to: when occlusion occurs in a person region, suspicious action determination processing for the person region in which occlusion occurs is prohibited.
In the crime prevention system according to the present disclosure, the processor may be configured to: when the recognition by the trace processing is changed to a failure in the middle of the transition, the suspicious activity determination processing is prohibited when the recognition by the trace processing is not continued for a predetermined time before the failure.
In the crime prevention system according to the present disclosure, the processor may be configured to further execute a notification process of notifying the alarm information from the output device when it is determined that the person of the suspicious action is included in the camera image.
In order to achieve the above object, the present disclosure provides a crime prevention method in which a computer performs monitoring of actions of persons included in a plurality of camera images continuously captured by a monitoring camera in accordance with a time series. The computer is configured to detect person regions of persons included in the plurality of camera images, perform a tracking process of recognizing persons included in the plurality of camera images in accordance with a time series based on person image data included in the person regions, and determine that a person whose action is suspicious is included in the camera images when the recognition based on the tracking process has failed in the middle of the transition.
In order to achieve the above object, the present disclosure is applied to a computer-readable recording medium having recorded thereon a crime prevention program for causing a computer to execute monitoring of actions of persons included in a plurality of camera images continuously captured by a monitoring camera in accordance with a time series. The crime prevention program is configured to cause a computer to execute: a person region of a person included in the plurality of camera images is detected, a tracking process is performed for recognizing the person included in the plurality of camera images in accordance with a time series based on the person image data included in the person region, and when the recognition based on the tracking process is changed to fail in the middle, it is determined that a person with suspicious action is included in the camera images.
According to the techniques related to this disclosure, suspicious actions contained in camera images of one or more surveillance cameras can be determined. Thereby, the camera images of one or more monitoring cameras can be used for crime prevention.
Drawings
Fig. 1 is a block diagram of a crime prevention system according to an embodiment of the present disclosure.
Fig. 2 is a block diagram showing an example of the structure of the crime prevention system.
Fig. 3 is a block diagram showing functions implemented by executing the crime prevention program 32 by the processor 20 of the management server 10.
Fig. 4 is a diagram showing a person region of person image data IMG1 detected from a camera image by a bounding BOX 1.
Fig. 5 is a diagram showing an example of a database of template image data.
Fig. 6 is a diagram for explaining the occlusion detection process.
Fig. 7 is a flowchart showing a routine of processing implemented by the crime prevention system.
(symbol description)
2: a monitoring camera; 4: an output device; 10: a management server; 20: a processor; 30: a memory; 32: crime prevention programs; 34: data; 100: a crime prevention system; 201: a tracking processing unit; 202: a reliability determination processing unit; 203: a shielding detection processing unit; 204: a suspicious action determination processing unit; 205: and a notification processing unit.
Detailed Description
Embodiments of the present disclosure are described below with reference to the drawings. However, in the embodiments described below, when the number, the amount, the range, and the like of each element are mentioned, the technical idea of the present disclosure is not limited to the mentioned number, except for the case where the number is specifically and clearly defined in principle. The structure and the like described in the embodiments shown below are not necessarily essential to the technical ideas according to the present disclosure, except for the case where they are particularly clearly shown and the case where they are clearly and principally specified.
Description of the embodiments
1. Summary of crime prevention System
Fig. 1 is a block diagram of a crime prevention system according to an embodiment of the present disclosure. The anti-crime system 100 determines suspicious persons from camera images of one or more surveillance cameras. The suspicious person here is a person who performs suspicious actions such as changing the appearance of clothing in a public place where a monitoring camera is installed.
The crime prevention system 100 includes one or more monitoring cameras 2, an output device 4 viewable by a manager, and a management server 10. The one or more monitoring cameras 2 are provided in, for example, a place where people pass through a public road or the like outside a commercial facility. One or more surveillance cameras 2 continuously take images of the passing population in time series. The type, number, and installation location of the monitoring cameras 2 are not limited. Fig. 1 illustrates an example of a camera image of a surveillance camera 2 that captures images of a crowd traveling on a road. Camera images captured by one or more monitoring cameras 2 are sent to the management server 10. The management server 10 may be directly connected to one or more monitoring cameras 2 or may be connected to one or more monitoring cameras 2 from a remote place via a communication network.
The management server 10 performs a "tracking process" of receiving camera images continuously captured in time series by one or more monitoring cameras 2 and tracking persons in the camera images. In the tracking process here, a technique called Person Re-Identification (pedestrian Re-recognition) is used, for example. In Person Re-Identification, first, a Person region with a Person is detected from an input camera image. Hereinafter, the image information of the person included in the detected person region is referred to as "person image data". The detected character image data is inputted as a search query to a neural network model for Person Re-Identification.
In addition, template image data of the candidate destination is input to the neural network model. The template image data is data in which identification IDs are associated with character image data identified in past camera images. In the neural network model, template image data of a candidate destination having the highest similarity to character image data input as a search query is output as an output result. In addition, when template image data of a candidate destination similar to the personal image data does not exist, data in which a new identification ID is associated with the personal image data is stored as a part of the new template image data. The management server 10 performs a Person Re-Identification-based tracking process on each of the inputted camera images in accordance with the time series, and recognizes a Person in accordance with the time series.
Here, the recognition of the personal image data based on the tracking process may be changed to failure in the middle. In the case of frequent failure such as the tracking process, first, the reliability reduction of the tracking process is considered. Such a situation may occur, for example, when an unsuitable image is stored in template image data as a candidate destination, when a camera image is unclear, when a person is an appearance that is difficult to recognize, or the like. The management server 10 is configured to be able to execute a "reliability determination process" of determining whether the reliability of the tracking process is lowered. In the reliability determination process, it is determined whether or not the tracking process for the subject personal image data continues for a predetermined time without failure.
In addition, when occlusion occurs in the tracking process, the tracking process may fail. The occlusion here refers to a situation in which a person in a person region cannot be accurately recognized because another object is superimposed on the foreground side of the person in the person region. The management server 10 is configured to be able to execute "occlusion detection processing" for detecting the occurrence of occlusion in tracking processing of the subject personal image data. In the occlusion detection process, for example, when the ratio of the foreground region of the personal image to the bounding box of the personal region is lower than a predetermined ratio, it is determined that occlusion has occurred.
Further, in the case where the appearance of the person suddenly changes in the middle of the tracking process, the tracking process may be changed to fail. Such a situation may arise, for example, in the case where a suspicious person is changed in order to evade tracking.
As described above, in the crime prevention system 100 according to the present embodiment, when the tracking process for the target person is executed with high reliability and no occlusion occurs, if the tracking process is changed to fail, it is determined that the target person has changed its appearance as a suspicious action. Hereinafter, this process will be referred to as "suspicious activity determination process".
When it is determined that the target person has performed a suspicious action, the management server 10 outputs alarm information from the output device 4 via the communication network. Hereinafter, this process will be referred to as "notification process". The output device 4 is exemplified by, for example, a monitor of a monitoring room displaying an image of the monitoring camera 2, a portable terminal of a monitor, and the like. Examples of the alarm information include a message for prompting attention to a suspicious activity, a sound notification, and the like.
According to the crime prevention system 100 as above, suspicious actions contained in the camera images of one or more monitoring cameras 2 can be determined. Thereby, the camera images of one or more monitoring cameras 2 can be used for crime prevention.
2. Structure of crime prevention system
Fig. 2 is a block diagram showing an example of the structure of the crime prevention system. The crime prevention system 100 includes one or more monitoring cameras 2, a management server 10, and an output device 4.
The management server 10 has a function as a processing device of a computer. Typically, the management server 10 is provided with one or more processors 20 (hereinafter simply referred to as processors 20) and one or more memories 30 (hereinafter simply referred to as memories 30) combined with the processors 20. In the memory 30, one or more crime prevention programs 32 (hereinafter, referred to simply as crime prevention programs 32) executable by the processor 20 and various data 34 associated with the neural network model 33 are stored.
By executing the crime prevention program 32 by the processor 20, various processes performed by the processor 20 are realized. Fig. 3 is a block diagram showing functions implemented by executing the crime prevention program 32 by the processor 20 of the management server 10. As shown in fig. 3, the processor 20 includes: a tracking processing section 201 for performing tracking processing; a reliability determination processing unit 202 for performing reliability determination processing; an occlusion detection processing unit 203 for performing occlusion detection processing; a suspicious activity determination processing unit 204 that determines suspicious activity included in the camera image; and a notification processing unit 205 for performing notification processing. The function of the processor 20 is described below with reference also to fig. 1.
The trace processing unit 201 is a functional block for performing trace processing based on Person Re-Identification. The tracking processing unit 201 detects personal image data from the camera image captured by the monitoring camera 2. In fig. 4, a person region of person image data IMG1 detected from a camera image is represented by a bounding BOX 1. The method of detecting the personal image data from the camera image can be applied to a known technique, and therefore, the description thereof will be omitted.
In the data 34 of the memory 30, a database of template image data is stored. Fig. 5 is a diagram showing an example of a database of template image data. The template image data is associated with an ID for identification with respect to the personal image data identified from the past camera image. The trace processing unit 201 has a neural network model 33 for Person Re-Identification. The neural network model 33 compares the personal image data with the database of template image data stored in the memory 30, and searches for a person having the same template image data as the personal image data, based on the similarity thereof. In addition, when template image data having a high similarity to the personal image data does not exist, an ID for identification is associated with the personal image data as new personal image data and registered in the database. The tracking processing unit 201 sequentially executes the above-described processing for a plurality of camera images that are continuous in time series, thereby tracking the person included in the camera images.
For example, in the case where the plurality of camera images includes a first camera image and a second camera image subsequent to the first camera image, in the tracking process for the second camera image, the person image data of the person region included in the second camera image is compared with the template image data. At this time, the template image data includes at least the personal image data of the personal area included in the first camera image. Therefore, in the tracking process for the second camera image, it is possible to recognize with the person image data of the person region included in the first camera image, whereby tracking of the person in accordance with the time series can be realized. The neural network model 33 used in the tracking process can be a well-known model for Person Re-Identification, and therefore, a detailed description thereof is omitted.
The reliability determination processing unit 202 is a functional block for performing reliability determination processing. In the reliability determination process, when the tracking process for the target person image fails, it is determined whether or not the tracking process before the failure has continued in a state of high reliability. The reliability index here is a duration for which the tracking process for the subject person image is continued without failure. The reliability determination processing unit 202 calculates a duration Tc before failure of the tracking process for the target person image, and compares the calculated duration Tc with a predetermined time Tth. Here, the predetermined time Tth is a value set in advance as a threshold value for determining that the reliability of the determination process is high. When the duration Tc is longer than the predetermined time Tth, the reliability determination processing unit 202 determines that the tracking process is continued in a highly reliable state.
The occlusion detection processing unit 203 is a functional block for performing occlusion detection processing. Fig. 6 is a diagram for explaining the occlusion detection process. Fig. 6 illustrates an example in which a part of the other personal image data IMG2 is superimposed on the foreground side of the personal area of the personal image data IMG1 surrounded by the bounding BOX 1. In the occlusion detection processing, the occlusion detection processing unit 203 calculates a ratio R of the foreground-side region of the personal image data IMG1 to the boundary BOX 1. Then, the occlusion detection processing unit 203 determines that occlusion has occurred when the calculated ratio R is greater than the predetermined ratio Rth.
The suspicious activity determination processing unit 204 is a functional block for executing suspicious activity determination processing. In the suspicious activity determination process, the suspicious activity determination process unit 204 determines that the person of the subject person image data has suspicious activity when the tracking process fails, the reliability determination process determines that the person is in a high reliability state, and the occurrence of occlusion is not detected in the occlusion detection process.
The notification processing portion 205 is a functional block for performing notification processing. In the notification process, the notification processor 205 outputs the alarm information from the output device 4 when the suspicious activity determination process has determined that there is a suspicious activity. The output method to the output device 4 is not limited. For example, the output mode may be to output a sound of attention notice from a speaker or to display an output such as pop-up to a monitor.
3. Specific use case for determining suspicious actions by crime prevention system
Next, an example of the process of determining suspicious actions by the crime prevention system 100 according to the embodiment will be described with reference to fig. 7. The crime prevention system 100 is a system that outputs an alarm for suspicious activity of a person in the case where suspicious activity made by the person is included in camera images IMG of one or more monitoring cameras. Fig. 7 is a flowchart showing a routine of processing implemented by the crime prevention system. The routine shown in fig. 7 is executed by the processor 20 of the management server 10 executing the crime prevention program 32. In addition, the flowchart also shows a part of the crime prevention method according to the embodiment of the present disclosure.
In step S100 of the routine shown in fig. 7, tracking processing is performed for the personal image data included in the camera images captured by the one or more monitoring cameras 2, respectively. In the next step S102, it is determined whether or not there is any person image data for which the tracking process in step S100 has been changed to failure. As a result, when the determination is not considered to be satisfied, the process of the present routine is ended, and when the determination is considered to be satisfied, the process proceeds to step S104.
In step S104, in the reliability determination process, it is determined whether or not the reliability of the tracking process before failure is high. In the reliability determination process, it is typically determined whether or not the duration Tc of the tracking process before failure is greater than the predetermined time Tth. As a result, when the determination is considered to be true, the process advances to step S106. On the other hand, if the determination is not considered to be satisfied, the process of the present routine is ended. Thereby, execution of suspicious action determination processing in subsequent steps is prohibited.
In step S106, it is determined whether or not occurrence of occlusion is detected in the occlusion detection process. As a result, when the determination is considered to be true, the process of the present routine is ended. Thereby, execution of suspicious action determination processing in subsequent steps is prohibited. On the other hand, if the determination is not considered to be satisfied, it is determined that suspicious activity is included in the camera image, and the process proceeds to step S108. In step S108, a notification process of notifying alarm information from the output device 4 is performed.
As is apparent from the above description, according to the crime prevention system 100 according to the embodiment, suspicious actions for performing appearance changes can be specified from camera images. As a result, the alarm information can be notified from the output device 4, and thus the alarm device is useful for crime prevention.
4. Modification examples
The crime prevention system 100 according to the embodiment may be modified as described below.
4-1 suspicious action determination processing
In the suspicious action determination process, it is not necessary to use the conditions of the occlusion determination process and the reliability determination process. That is, in the suspicious activity determination process, if it is considered that the suspicious activity determination process is satisfied in step S102, the process may proceed to step S108 without performing the process of step S104 and the process of step S106.
4-2 tracking process
The tracking process is not limited in its way if a person can be recognized between camera images in accordance with a time series.

Claims (7)

1. A crime prevention system that monitors actions of persons included in a plurality of camera images continuously captured in time series by a monitoring camera, the crime prevention system comprising:
a memory storing the plurality of camera images; and
a processor for processing the plurality of camera images stored in the memory,
the processor is configured to:
tracking processing of detecting person regions of persons included in the plurality of camera images, respectively, and recognizing the persons included in the plurality of camera images in accordance with a time series based on the person image data included in the person regions; and
and suspicious action determination processing for determining that a person of a suspicious action is included in the camera image when the recognition based on the tracking processing is changed to fail in the middle.
2. The crime prevention system of claim 1, wherein,
the plurality of camera images includes a first camera image and a second camera image subsequent to the first camera image,
in the memory, template image data including character image data of the character region of the first camera image is stored,
the tracking process is configured to recognize a person of the second camera image by comparing person image data included in the person region of the second camera image with the template image data using a neural network model.
3. The crime prevention system of claim 1, wherein,
the processor is configured to prohibit the suspicious action determination process for the person region in which the occlusion occurs, when the occlusion occurs for the person region.
4. A crime prevention system according to any one of claims 1 to 3 wherein,
the processor is configured to prohibit the suspicious action determination process when the recognition based on the tracking process is not continued for a predetermined time before the failure in a case where the recognition based on the tracking process is changed to a failure in the middle.
5. The crime prevention system of any of claims 1 to 4, wherein,
the processor is configured to further execute a notification process of notifying alarm information from an output device when it is determined that the person of the suspicious activity is included in the camera image.
6. A crime prevention method, a computer performing monitoring of actions of persons contained in a plurality of camera images continuously captured in time series by a monitoring camera, wherein the crime prevention method comprises:
detecting person areas of the person included in the plurality of camera images, respectively;
performing a tracking process of recognizing a person included in the camera image in accordance with a time series from the person image data included in the person region; and
when the recognition based on the tracking process is changed to fail in the middle, the person determined to be a suspicious action is included in the plurality of camera images.
7. A computer-readable recording medium configured to record a crime prevention program that causes a computer to execute monitoring of actions of persons included in a plurality of camera images continuously captured by a monitoring camera in accordance with a time series, wherein,
the crime prevention program causes a computer to execute:
detecting person areas of the person included in the plurality of camera images, respectively;
performing a tracking process of recognizing persons included in the plurality of camera images in accordance with a time series from the person image data included in the person region; and
when the recognition based on the tracking process is changed to fail in the middle, the person determined to be a suspicious action is included in the camera image.
CN202211577541.1A 2021-12-10 2022-12-09 Crime prevention system, crime prevention method, and computer readable recording medium Pending CN116259000A (en)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2021201005A JP2023086471A (en) 2021-12-10 2021-12-10 Crime prevention system, crime prevention method, and crime prevention program
JP2021-201005 2021-12-10

Publications (1)

Publication Number Publication Date
CN116259000A true CN116259000A (en) 2023-06-13

Family

ID=86680057

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202211577541.1A Pending CN116259000A (en) 2021-12-10 2022-12-09 Crime prevention system, crime prevention method, and computer readable recording medium

Country Status (3)

Country Link
US (1) US20230186741A1 (en)
JP (1) JP2023086471A (en)
CN (1) CN116259000A (en)

Family Cites Families (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JPWO2015166612A1 (en) * 2014-04-28 2017-04-20 日本電気株式会社 Video analysis system, video analysis method, and video analysis program
JP6729793B2 (en) * 2017-03-30 2020-07-22 日本電気株式会社 Information processing apparatus, control method, and program
US11062579B2 (en) * 2019-09-09 2021-07-13 Honeywell International Inc. Video monitoring system with privacy features
US11783652B2 (en) * 2020-06-15 2023-10-10 Honeywell International Inc. Occupant health monitoring for buildings

Also Published As

Publication number Publication date
US20230186741A1 (en) 2023-06-15
JP2023086471A (en) 2023-06-22

Similar Documents

Publication Publication Date Title
JP4924607B2 (en) Suspicious behavior detection apparatus and method, program, and recording medium
US10346688B2 (en) Congestion-state-monitoring system
US7796780B2 (en) Target detection and tracking from overhead video streams
JP5560397B2 (en) Autonomous crime prevention alert system and autonomous crime prevention alert method
US20180115749A1 (en) Surveillance system and surveillance method
US20050271250A1 (en) Intelligent event determination and notification in a surveillance system
JP6729793B2 (en) Information processing apparatus, control method, and program
KR101858396B1 (en) Intelligent intrusion detection system
KR102356666B1 (en) Method and apparatus for risk detection, prediction, and its correspondence for public safety based on multiple complex information
JP2008108243A (en) Person recognition device and person recognition method
KR102263512B1 (en) IoT integrated intelligent video analysis platform system capable of smart object recognition
KR101979375B1 (en) Method of predicting object behavior of surveillance video
JP5202419B2 (en) Security system and security method
Gavaskar et al. A novel design and implementation of IoT based real-time ATM surveillance and security system
JP5752977B2 (en) Image monitoring device
CN104050785A (en) Safety alert method based on virtualized boundary and face recognition technology
JP2024009906A (en) Monitoring device, monitoring method, and program
CN116259000A (en) Crime prevention system, crime prevention method, and computer readable recording medium
KR20030040434A (en) Vision based method and apparatus for detecting an event requiring assistance or documentation
JP2020191585A (en) Information processing device, information processing method, and program
CN111104845A (en) Detection apparatus, control method, and computer-readable recording medium
JP7347481B2 (en) Information processing device, information processing method, and program
JP2019153908A (en) Image authentication system, image authentication method and image authentication program
KR102286675B1 (en) Method for implementing security system based on motion recognition and security system using the method
KR20230100242A (en) Ai image analizing method for pereimeter intrusion detection system using cctv

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination